본문 바로가기 주메뉴 바로가기
22대 대한민국 국회 국회정보길라잡이 국회의원검색
회원가입 로그인
HOME

정보검색

소장정보 검색

국회도서관 홈으로 정보검색 소장정보 검색

결과 내 검색

동의어 포함

고급검색

인명/단체명

저자정보 상세정보
인명/단체명을 입력하세요.

키워드

  • 대표어

  • 외국어

-

목차보기

Title Page 2

Contents 5

Abstract 14

Chapter 1. INTRODUCTION 16

1.1. Background 16

1.1.1. Robot-aided therapy 18

1.1.2. Toward effective robot-aided therapy 19

1.1.3. Challenges 21

1.2. Literature Survey 23

1.2.1. Overview of Individualized Training Frameworks 23

1.2.2. Individualized Mapping Performance 25

1.2.3. Robot training with vector field 26

1.2.4. Motor Improvement Tracking Method 27

1.2.5. Balancing the difficulty mechanism 27

1.3. Objectives of Research 30

1.4. Problem Statements 30

1.4.1. Model-based Evaluation 30

1.4.2. Training Scheduling Method 32

Chapter 2. Model-based Evaluation 35

2.1. Introduction 35

2.2. Individually scaled evaluation method 36

2.2.1. Model formulation for normal reaching movement 37

2.2.2. Evaluation method based on normal reaching model 39

2.2.3. Visualization of evaluation results 42

2.3. Experiments 42

2.3.1. Experimental design 42

2.3.2. Participants 46

2.3.3. Protocols 47

2.3.4. Data analysis 48

2.4. Results 51

2.4.1. Normal Reaching Model Performance 51

2.4.2. Evaluation Visualization 61

2.4.3. Discussion 64

2.4.4. Conclusion 68

Chapter 3. Individualized Training Framework 70

3.1. Introduction 70

3.1.1. Proposed Framework 71

3.2. Theoretical: Simulation of scheduling method 71

3.2.1. Simulation Architecture 71

3.2.2. Reaching Training Environment 72

3.2.3. Recovery Model 72

3.2.4. Training Scheduling 75

3.2.5. Data Collection and Virtual Patients 76

3.2.6. Data Analysis 77

3.2.7. Result 78

3.2.8. Discussion 80

3.3. Practical: Scheduling method accounting for evaluation validity 84

3.3.1. Introduction 84

3.3.2. Method 85

3.3.3. Preliminary Test and Discussion 91

Chapter 4. Model Modification 95

4.1. Introduction 95

4.2. Model Exploration 97

4.2.1. Feature Exploration 97

4.2.2. Data Preparation 98

4.2.3. Exhaustive Search of Non-linear Regression Model 98

4.2.4. Data Selection Method 106

4.2.5. Data Analysis 108

4.3. Result 110

4.3.1. Kinematic Features Explaining Healthy Reaching Movement Time 110

4.3.2. Candidate Models in Healthy Reaching Data 112

4.3.3. Candidate Models in Stroke Reaching Data 113

4.3.4. Individualized Training Simulation 115

4.4. Discussion 117

Chapter 5. Conclusion 122

5.1. Summary 122

5.2. Limitations 124

Bibliography 127

초록 145

List of Tables 12

Table 2.1. Participants details of affected arm in validation experiments 46

Table 2.2. Participants details of affected arm in pilot study 49

Table 2.3. Summary of experiment configuration 49

Table 2.4. AIC and R2 values of candidate models in validation experiments 53

Table 2.5. AIC and R2 values of candidate models in pilot study 55

Table 2.6. R² values of the proposed model and the average kinematic data for each post-stroke 58

Table 4.1. List of features related to loss of somatosensation 99

Table 4.2. List of features related to paresis 100

Table 4.3. List of candidate features derived from literature review, grouped by movement characteristics and symptom loss... 101

Table 4.4. List of candidate features derived from literature review, grouped by loss of abnormal muscle tone category 102

Table 4.5. List of candidate features derived from literature review, grouped by loss of fractionated movement category 103

Table 4.6. List of candidate features derived from literature review, grouped by miscellaneous category 104

Table 4.7. List of assorted candidate features by overall feasibility 105

Table 4.8. Comparison of candidate models for normal reaching using healthy reaching data 112

Table 4.9. Statistical results of candidate models for the normal reaching model. sig. indicates the significance level of the... 114

List of Figures 9

Figure 1.1. Reaching training on rebless.planar (commercialized rehabilitation robot.) 17

Figure 1.2. Virtual environment of reaching task. Yellow cursor represents the hand position of... 19

Figure 1.3. Motor learning principles 20

Figure 1.4. Reaching evaluation validity 22

Figure 1.5. Individualized mapping method 26

Figure 1.6. Motor distribution of stroke patients 28

Figure 1.7. Physical representation of index of difficulty in Fitts's Law 29

Figure 1.8. Proposed framework. (a) is a simplified block diagram of the framework flow. (b)... 34

Figure 2.1. Development process of individually scaled evaluation method 36

Figure 2.2. Example linear regression fit for a reaching model 41

Figure 2.3. Example assessment profile mapping 41

Figure 2.4. Experimental setup for validation experiments 44

Figure 2.5. Experimental setup for the pilot study 45

Figure 2.6. Comparison of average Akaike information criterion (AIC) for the candidate models 52

Figure 2.7. Comparison of model R². The boxplot represents the distribution, and the scatter plot... 56

Figure 2.8. Relationship between residuals of each candidate model and erroneous reaching pa-... 59

Figure 2.9. Relationship between residuals of each candidate model and erroneous reaching pa-... 60

Figure 2.10. Reaching profile of post-stroke participants for the affected sides in the validation... 62

Figure 2.11. Reaching profile of post-stroke participants for affected sides in the pilot study 63

Figure 2.12. R² distribution of the proposed model in healthy and less-affected conditions 66

Figure 3.1. Theoretical and practical level of training framework 71

Figure 3.2. Comparison of simulated recovery amount 79

Figure 3.3. Composite spatial-temporal entropy of reaching trajectories for different scheduling... 80

Figure 3.4. Example training visualization of virtual patients 81

Figure 3.5. Comparison of total recovery amount 83

Figure 3.6. Effect of robot assistance on time duration of reaching 85

Figure 3.7. Overall schematic of the quasi-assessment method considering assisted movements and... 86

Figure 3.8. Algorithm of the quasi-assessment method 87

Figure 3.9. Overall schematic of the implementation process for the individualized reaching training... 88

Figure 3.10. Protocol for the preliminary test for the proposed framework 90

Figure 3.11. Sequence of the proposed framework considering the adaptive training and quasi-... 92

Figure 3.12. Interaction force during adaptive training and quasi-assessment trial 93

Figure 4.1. Illustration of normal trial selection based on kinematic feature space us-ing the DBSCAN... 107

Figure 4.2. Feature importance based on Shapley values for predicting movement time during... 111

Figure 4.3. Comparison of simulated recovery amount with the modified model 115

Figure 4.4. Composite spatial-temporal entropy of reaching trajectories for different scheduling... 116

Figure 4.5. Correlation of the normalized errors and the kinematic variables 120

초록보기

 상지 재활 로봇은 뇌졸중 환자에게 반복적인 도달 운동 훈련을 제공할 수 있으나 초기 로봇 재활 시스템은 단순한 도달 환경과 획일화 된 스케줄만을 제공하기 때문에 환자 개인의 특성을 반영하지 못하였다. 본 학위논문은 모델 기반 평가법을 활용하여 환자별 맞춤형 도달 훈련 스케줄을 자동 생성하는 개인화 도달 운동 훈련 프레임워크를 제안한다. 먼저, 시간적·공간적 편향 없이 작업공간 전반의 도달성능을 평가할 수 있는 정량적 평가 기법을 개발하였다. 해당 기법은 1. Fitts 법칙 (속도–정확도), 2. Almanji 모델 (뇌성마비 대상 마우스 포인팅 동작), 3. 본 연구의 제안 모델 등 세 가지 후보 모델을 비교 하고자. 모델 검증을 위해 건강인 12 명, 뇌졸중 환자 7 명으로부터 로봇 기반 운동학 데이터를 수집·분석하였다. 건측의 데이터를 이용해 정상 도달 모델을 추정하고, 이를 기준으로 환측 팔을 평가하도록 설계하였다. 제안 모델은 건강인 12 명 전원 및 뇌졸중 환자 19 명의 덜 손상된 팔 중 16 명에서 R² > 0.7 이상의 설명력을 보였으나, 환측 팔의 비정상 궤적은 올바르게 식별하지 못해 평가 지표로서의 정상 움직임에 대한 특이성 (specificity) 을 입증하였다. 시각화된 평가 결과는 환측 팔의 고유 운동 특성을 직관적으로 제시하여 맞춤형 훈련 과제 선정에 활용 가능함을 확인하였다. 이후 운동 학습(motor-learning) 원리를 반영한 스케줄링 알고리즘을 제안하고. 알고리즘은 적응성 (adaptiveness)과 무작위성 (randomness) 을 동시에 반영해 회복 촉진과 기술 유지 (skill retention) 를 도모한다. 실제 환자 데이터를 기반으로 한 컴퓨터 시뮬레이션 결과, 제안 스케줄은 기존 방법 대비 훈련 적응도 및 변동성에서 유의미한 우수성을 보였다 (p < 0.001). 또한 기존 정상 도달 운동 모델 고도화를 위해 건강인 및 뇌졸중 환자의 광범위한 기구학적 변수를 분석하였다. 교차검증과 AIC 등을 활용해 비선형 회귀모델을 탐색하여, 정상 도달 시간 예측에 가장 영향력 있는 변수 집합을 도출하였다. 개선된 모델은 기존 접근법과 비교해 통계적 설명력과 특이성이 향상되었음을 검증하였다.

추천서가 (다양한 추천 자료를 만나보세요)

권호기사보기

권호기사 목록 테이블로 기사명, 저자명, 페이지, 원문, 기사목차 순으로 되어있습니다.
기사명 저자명 페이지 원문 기사목차

권호

기사명 원문보기 기사목차
닫기